SI
SI
discoversearch

We've detected that you're using an ad content blocking browser plug-in or feature. Ads provide a critical source of revenue to the continued operation of Silicon Investor.  We ask that you disable ad blocking while on Silicon Investor in the best interests of our community.  If you are not using an ad blocker but are still receiving this message, make sure your browser's tracking protection is set to the 'standard' level.
Strategies & Market Trends : Technical Analysis - Beginners

 Public ReplyPrvt ReplyMark as Last ReadFilePrevious 10Next 10PreviousNext  
To: CatLady who wrote (10379)5/31/1999 12:51:00 AM
From: keith massey  Read Replies (4) of 12039
 
Thanks Catlady

I'm presently doing my P.hd in Neurophysiology. One of the main things I do in my research is find patterns in complex electrical waveforms (Electromyography (EMG), ECG, EEG, etc.). When first observing these waveforms they appear like a bunch of random signal (noise) with no pattern (kind of like a stock chart). The first thing I do is apply a 4 or 6 pole low or high pass recursive filter (or bandpassed). Then by using waveform triggering, averaging, regression, etc., I can start to find patterns in these waveforms which can be used for predictions. You can understand why I was drawn to technical analysis of stock charts....it was like I have been in training for years to do this.

Everyone uses moving averages to smooth out the waveforms. However the main problem with moving averages it that the waveform is time shifted and this time shift is dependent on the days averaged. Simply time shifting the moving average doesn't solve the problem since you are missing the front end data. By using a recursive filter the waveform is not time shifted and the wave represent the true underlying movement of the stock.

I'm not sure if this has been done before but my plan is to use two or three high pass filters with different cut off frequencies and overlay them on a chart. In theory this should give the best prediction of the underlying movement of a stock trend. It works in the electophysiology field <gggg>

I have read 25-30 TA books and have never heard of this approach yet. I'm sure someone has done it but I haven't seen the book or writeup yet (some please give me a reference). The big problem now is to write the code that can filter a waveform in Metastock. I actually imported one stock as a text file into my analysis software in the lab and performed the filtering. It worked great. However this software is to clumsy to use on a database of stocks. I can't use the code from my lab software because it was written in C++.

Anyone got any suggestions.

Best Regards
KEITH
Report TOU ViolationShare This Post
 Public ReplyPrvt ReplyMark as Last ReadFilePrevious 10Next 10PreviousNext